[PATCH v6 6/6] drm/xe/xe_vm: Implement xe_vm_get_faults_ioctl

[Jonathan Cavitt <jonathan.cavitt@xxxxxxxxx>]

Add support for userspace to request a list of observed failed
pagefaults from a specified VM.

v2:
- Only allow querying of failed pagefaults (Matt Brost)

v3:
- Remove unnecessary size parameter from helper function, as it
  is a property of the arguments. (jcavitt)
- Remove unnecessary copy_from_user (Jainxun)
- Set address_precision to 1 (Jainxun)
- Report max size instead of dynamic size for memory allocation
  purposes.  Total memory usage is reported separately.

v4:
- Return int from xe_vm_get_property_size (Shuicheng)
- Fix memory leak (Shuicheng)
- Remove unnecessary size variable (jcavitt)

v5:
- Rename ioctl to xe_vm_get_faults_ioctl (jcavitt)

Signed-off-by: Jonathan Cavitt <jonathan.cavitt@xxxxxxxxx>
Suggested-by: Matthew Brost <matthew.brost@xxxxxxxxx>
CC: Jainxun Zhang <jianxun.zhang@xxxxxxxxx>
CC: Shuicheng Lin <shuicheng.lin@xxxxxxxxx>
---
 drivers/gpu/drm/xe/xe_device.c |  3 ++
 drivers/gpu/drm/xe/xe_vm.c     | 78 ++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_vm.h     |  2 +
 3 files changed, 83 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 5d79b439dd62..2d22b3c2df09 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -194,6 +194,9 @@ static const struct drm_ioctl_desc xe_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(XE_WAIT_USER_FENCE, xe_wait_user_fence_ioctl,
 			  DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(XE_OBSERVATION, xe_observation_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(XE_VM_GET_FAULTS, xe_vm_get_faults_ioctl,
+			  DRM_RENDER_ALLOW),
+
 };
 
 static long xe_drm_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index eada3ecc2364..e24601707f2a 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -3538,6 +3538,84 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	return err;
 }
 
+static int xe_vm_get_faults_size(struct xe_vm *vm)
+{
+	int size;
+
+	spin_lock(&vm->pfs.lock);
+	size = vm->pfs.len * sizeof(struct xe_vm_fault);
+	spin_unlock(&vm->pfs.lock);
+
+	return size;
+}
+
+static int fill_faults(struct xe_vm *vm,
+		       struct drm_xe_vm_get_faults *args)
+{
+	struct xe_vm_fault __user *usr_ptr = u64_to_user_ptr(args->faults);
+	struct xe_vm_pf_entry *entry;
+	int ret = 0, i = 0;
+
+	spin_lock(&vm->pfs.lock);
+	list_for_each_entry(entry, &vm->pfs.list, list) {
+		struct xe_pagefault *pf = entry->pf;
+
+		ret = put_user(pf->page_addr, &usr_ptr->address);
+		if (ret)
+			break;
+
+		ret = put_user(pf->address_type, &usr_ptr->address_type);
+		if (ret)
+			break;
+
+		ret = put_user(1, &usr_ptr->address_precision);
+		if (ret)
+			break;
+
+		usr_ptr++;
+
+		if (i == args->fault_count)
+			break;
+	}
+	spin_unlock(&vm->pfs.lock);
+
+	return ret ? -EFAULT : 0;
+}
+
+int xe_vm_get_faults_ioctl(struct drm_device *drm, void *data,
+			   struct drm_file *file)
+{
+	struct xe_device *xe = to_xe_device(drm);
+	struct xe_file *xef = to_xe_file(file);
+	struct drm_xe_vm_get_faults *args = data;
+	struct xe_vm *vm;
+	int size, fault_count;
+
+	if (XE_IOCTL_DBG(xe, args->reserved[0] || args->reserved[1]))
+		return -EINVAL;
+
+	vm = xe_vm_lookup(xef, args->vm_id);
+	if (XE_IOCTL_DBG(xe, !vm))
+		return -ENOENT;
+
+	size = xe_vm_get_faults_size(vm);
+	fault_count = size / sizeof(struct xe_vm_fault);
+
+	if (size < 0) {
+		return size;
+	} else if (!args->size && !args->fault_count) {
+		args->size = size;
+		args->fault_count = fault_count;
+		return 0;
+	} else if (args->size > size || args->fault_count > fault_count) {
+		return -EINVAL;
+	} else if (args->size / sizeof(struct xe_vm_fault) != args->fault_count) {
+		return -EINVAL;
+	}
+
+	return fill_faults(vm, args);
+}
+
 /**
  * xe_vm_bind_kernel_bo - bind a kernel BO to a VM
  * @vm: VM to bind the BO to
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index fe068e4448a6..c48bd6e6fe34 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -191,6 +191,8 @@ int xe_vm_destroy_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file);
 int xe_vm_bind_ioctl(struct drm_device *dev, void *data,
 		     struct drm_file *file);
+int xe_vm_get_faults_ioctl(struct drm_device *dev, void *data,
+			   struct drm_file *file);
 
 void xe_vm_close_and_put(struct xe_vm *vm);
 
-- 
2.43.0